Modeling and driving a reduced human mannequin through motion captured data: A neural network approach

One of the major problems which arises in the field of virtual design is th e realization of virtual mannequins able to move in a human like way. The m annequin represents a fundamental part of the whole computer-aided (CAD) sy stem, being the virtual environments nowadays very well described. This wor k is focused on the analysis of the human sitting working posture, which is described by a 30 degree of freedom (DOF) mannequin, modeling the upper pa rt of the body (pelvis, trunk, arms, and head). Trajectories formation in p oint to point reaching movements represents the main topic, Our approach is based on the acquisition of real human kinematics data, collected by means of an automatic motion analyzer. Starting from the kinematics database of one subject, sit in front of a desk, a neural network was trained in order to generate the movements of the virtual mannequin. The work is divided int o four parts: mannequin modeling, three-dimensional (3-D) human data collec tion, data preprocessing according to the biomechanical model, and design a nd training of a multilayer perceptron neural network.